head crashes into bed on movement during large prints.

[Qdeathstar]

I have been printing a lot of things around 6-8" without an issue, but twice now when I print large things the head crashes into the bed causing motors to skip steps and the head to end up outside the printable area. it seems to happen on movement (not while laying down filament).


it doesn't happen during smaller prints. I was printing the Rostock calibration test print when it happened, the test had printed a brim and the first shell of the part before the issue happened, and those lines looked good.


I'm wondering if the belt tension is wrong and the belt is slipping while trying to pull the arms back up from a almost fully extended position....

[Jimustanguitar]

If it's intermittent and not 100% repeatable, I'd suspect that you have a loose endstop switch. Something mechanical is changing from one print to another.

Are you printing the same material and running the bed at the same heat each time? That could cause a little bit of a Z shift as well. Zero your Z height for each different bed temp as a precaution.

[nebbian]

I've occasionally had similar issues, where the head crashes into a part of the print that has curled up slightly. This causes a skipped step, and consequent failure.

I've found that setting z hop (the height that the nozzle raises while travelling) to around 0.4mm helps a lot with this sort of issue. Deltas handle Z hop really well, unlike many cartesian printers with a slow Z axis.

[DeltaCon]

If it's intermittent and not 100% repeatable, I'd suspect that you have a loose endstop switch. Something mechanical is changing from one print to another.

I can see why an endstop problem would cause problems during homing. But how would that cause problems like this during mid-print? I think most of these problems come down to either under or over current on the xyz steppers that causes missed steps (by too low torque or overheated stepperdrivers), or too high speeds and acceleration settings that the steppers cannot live up to.
So check the brand of your steppers and change the motor current setting in your firmware, and try a lower travelspeed.

[Mac The Knife]

When I ran into this problem on my printer, I was trying to print a Savonius turbine, and was trying to print out to the maximum diameter of the build area. It would fail as you describe, in the same spot. After ruling out mechanical and electrical possibilities, I chalked it up to the limitations of the RamBo board.

[Jimustanguitar]

I can see why an endstop problem would cause problems during homing. But how would that cause problems like this during mid-print?

If the print starts out great, and then trouble shows up mid way through, you're exactly right. I wasn't sure one way or the other from the original post.


If you think this is happening on travel moves, you can certainly slow them down. Moving too fast can easily skip steps. I usually set non-printing moves around 150mm/s. If you're still having trouble, kick it down to 100 as a test.

How are your motor currents? Are the motors hot, but not too hot to touch? If your currents are too high or too low, you'll not be able to move as quickly without skipping steps.

[Qdeathstar]

I used the settings recommended by gene. What happens was it printed a single line around and then as it positioned itself to start printing the inner portion it hit the plate. I was watching it (knowing I had the issue before" but it happens so fast I'm not sure it it hit and drug (the filament on the glass didn't look dinged) or it it just fell.

It seems like it might be a z-axis issue though, because I am currently printing the x-axis peice of the Rostock v2 enclosure which goes from the y-axis pole to the z-axis pole almost the full width and it is printing without issue. And the problem happed yesterday with the z arm fully stretched down to the bed.

[Jimustanguitar]

It seems like it might be a z-axis issue though, because I am currently printing the x-axis peice of the Rostock v2 enclosure which goes from the y-axis pole to the z-axis pole almost the full width and it is printing without issue. And the problem happed yesterday with the z arm fully stretched down to the bed.

You may also want to check if the cheapskate bearing is out of travel and hitting the base of the printer. Depending on which generation of the carriages you have, clearance is tight at the extreme edge.

[Qdeathstar]

I have the newest version cheapskates, but it's possible. I printed out some bed clamps that might reduce clearance

[geneb]

Question - when this happens do one of the carriages reach a certain point and just kind of drop to the bottom of the tower (bottom of the carriage is resting on the top of the base)?

g.

[guanu]

change your non print travel speed down, if it is set at 300 (which we have asked for the default to change, but it hasnt yet) bring it down to about 150 and that issue will go away.

[Qdeathstar]

Ok, I will do that.

@geene, it starts moving then drops, the head is 1mm off the bed(bad calibration) so it's not a issue with the nozzle hitting first.

I got a video I'm in the process of uploading..


Why would the nozzle be 1mm off the bed at the front but much closer to true everywhere else.

[geneb]

Yeah, what happens is that unless it goes pretty slow at the outside edge, the carriage can drop. Are you using the ball & socket arms with the new carriages or are you still using the stock arms?

tnx.

g.

[Qdeathstar]

New carriage

[geneb]

I asked about the arms. I already knew you had the new carriages.

g.

[Qdeathstar]

new arms, i think. They are plastic. I bought the machine in april from MatterHackers.

[Qdeathstar]

http://forum.seemecnc.com/viewtopic.php ... =25#p89725

^picture of the machine

The video was kind of shaky/probably not worth the time. and i started another print.

[geneb]

They're actually glass-filled nylon.

Yeah, I suspect your movement speed is what's getting you.

g.

[Qdeathstar]

I'm printing for reals right now, but I'll change a few settings and see what happens.

My Prussa needed a lot of changes to the settings so I was in there a lot of time. The Rostock printed good enough out the box so I havnt messed with anything out of fear of messing something up.

[Qdeathstar]

Looks like lowering the speed to 150mm/s solved this issue. Thanks!

[geneb]

Qdeathstar, now that you've got it working, it's tweaking time! Bump your rapid speed up by 25mm/s and re-run your problematic job. Keep doing that until it fails then back off 25mm/s. That'll give you the most bang for the buck.

g.

[Tonkabot]

Just checking in - it's been a while. I am having this problem too - I had been running from a laptop, but I brought the printer into work where it has a dedicated computer. So I had to install Mattercontrol from scratch and use with the default Rostock Max V2 settings.
When mine fails, only on large prints (8"x 4" base), the head even starts twisting and doing goofy things.
It also seems to have failed before it even starts to print the skirt, it makes goofy and slow moves getting down to the printbed, which it doesn't even reach in some cases.

I have the Rambo, cheepskates, tricklaser arms, and 400 step/rev steppers instead of the stock 200 step/rev.

Printing stuff the size of a raspberry pi case or smaller work great with no problems.

So just to be sure, is this speed setting in the EEPROM, or is it a printer setting? Or does it sound like something else?

[Qdeathstar]

It was a print setting for me, I changed it on the matter control tablet.

You problem sounds different than mine. With my problem, it would print the outer shell of the first layer fine, then when it jumped to the infill it would crash.

I think you need to change your print radius and arm length. I don't the the trick later arms are the same length as a stock Romax.